W3C home > Mailing lists > Public > www-tag@w3.org > September 2007

Can a URI be a resource? If not, the URIs can't identify EVERYTHING

From: Nick Gall <nick.gall@gmail.com>
Date: Tue, 25 Sep 2007 21:37:39 -0400
Message-ID: <a6a6df360709251837l4cc5b8f0s4078a2aff928086b@mail.gmail.com>
To: www-tag@w3.org
It's 2:30am in London where I write this, just having read the umpteenth
post about Cool URIs, so I may just be halucinating. But it struck me when
thinking about the things that a URI can't identify that AFAIK a URI can't
identify a URI. Is this true? If so, is it important?

It could be important because without a way to identify a URI, one cannot
make (RDF) assertions it. For example, given the http URI

[1] http://ironick.typepad.com/ironick/2007/09/three-kinds-of-.html

one might want to assert that it "ends with the letters h,t,m,l" or that it
is so many characters long.

If a URI can identify a URI, then what would the URI for [1] look like, eg

http://uri.reference.org/?uri=http://ironick.typepad.com/ironick/2007/09/three-kinds-of-.html


And what content-type would be used for the representation returned? text?

I raise this issue, not only because I am curious, but also because I think
it sheds light on the issue of URI ambiguity being discussed in the Cool
URIs permathread.

Just as the argument can be made that we must be able to speak about and
unambiguously identify "information resources" independent of
"non-information resources" (things in themselves), the argument can be made
that we must be able to speak about and unambiguously identify "identifiers"
("identifying resources" perhaps) independent of the things they identify.
In English, we simply use quotation marks to identify the identifier, eg
"'Nick' is a nickname of Nicholas Gall."

So for the person Alice, we need at least URIs for three entities to be
identified:

   1. A URI identifying Alice herself: http://www.acme.com/id/alice
   2. A URI identifying an information resource representing Alice:
   http://www.acme.com/data/alice
   3. A URI identifying [1], in other words a URI identifying the URI
   that identifies Alice: http://www.acme.com/?uri=
   http://www.acme.com/id/alice

Since we've gotten along without (3) so far, perhaps we can get along
without (1) as well. Just a thought.

In any case, I like the idea that URIs seem to identify everything that can
be thought--except themselves.

-- Nick

-- 
Nick Gall
Phone: +1.781.608.5871
AOL IM: Nicholas Gall
Yahoo IM: nick_gall_1117
MSN IM: (same as email)
Google Talk: (same as email)
Email: nick.gall AT-SIGN gmail DOT com
Weblog: http://ironick.typepad.com/ironick/
Furl: http://www.furl.net/members/ngall
Received on Wednesday, 26 September 2007 01:37:54 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 22:56:18 UTC